Saturday, October 07, 2017

The Killer Rabett of Caerbannog Has a Chew on Steve (Tony Heller) Goddard

Eli's friend, the Killer Rabett of Caerbannog noticed Eli's comment on Steve Tony Goddard's sense of misdirection, and had a few words on the Twitter.  There were more that a few clues to Steve Tony's methods and materials which, both of us agree, are worthy of gathering in one place for future use with a few expansions of Twitterisms and such.  Rabett of caerbannog T-shirts available at Shirtoid

In response to the usual cherry, pick with a topping of political agenda, Caerbannog. . . well let Eli go to the Twitter

Here, says Caerbannog pointing to Eli, is proof that Goddard aka Heller is a tinfoil-hat-wearing nutter

What else can be said about someone who works a construction project into a grand global conspiracy to manipulate temperatures? He's *nuts*.

As for the content of the message? I've addressed data adjustment issues many times here, but you fail to understand.

  • I have shown many times that adjustments are required to correct for the effects of station moves (i.e. 100s from city centers to airports). 
  • I have pointed you to information about data adjustmnts, including links to NOAA's adjustment package that includes all code
  • *and* full instructions on how to build/run it. The NOAA package includes a complete test-suite to make it easier for others to verify....
Bottom line -- when I post substance, you completely ignore it and blow it off.

Recap for others here: Heller focuses on USHCN instead of GHCN because USHCN covers only about 1.5% of the Earth's surface area. As a result, short term temp swings (aka "noise") are  highly correlated over temp stations. When averaged together, this results in less "noise cancellation".

That means a lower "warming" SNR. The US-only temperature trend is just emerging. A small local trend means adjustments can change slight cooling to slight warming. Also, there is more likely to be a temperature bias for localized regions, where many temperature stations may have had similar changes in equipment and procedures.

It turns out that for USHCN as well as GHCN, many stations were moved from city centers to airports. Many airport stations have data from before 1900.

When you see airport stations with data going back before there were airplanes, what does that tell you about those stations? Think hard.

Also warming has not been uniform over the continental USA. The West has warmed more than the South. That means area weighting is important.

Overall station density a bit less in the West than in other regions. So without area-weighting, the area with the most warming is underrepresented.

The problem can be worsened if you select a subset of stations. You must ensure reasonably uniform geographic weighting in your averaging. This is especially true if the warming is fairly weak relative to the year-to-year variation. This will often be true for small regions.

USHCN, because it covers only 1.5% of Earth area with annual temp variations large relative to warming, gives Heller opportunities to mislead. If Heller were to look at GHCN, where coverage is global and the warming trend is much larger than global year-to-year variations he wouldn't have any "wiggle room" to spin results and make accusations. The global temperature trend is huge relative to the noise and huge relative to adjustments.

That's why Heller won't touch GHCN. The warming signal to noise ratio for the entire Earth is much stronger than the warming signal to noise ratio for just the USA.

There's no way that Heller could "cherry pick" the GHCN warming trend away. The warming signal too strong. So Heller will never touch GHCN.

So Heller will never touch GHCN. That is, Heller will *never* show adjusted vs. unadjusted *global* temp averages from GHCN data. The results would destroy his message.



So I've noticed:

steven mosher said...

ya, I've mentioned this a bunch of times..

caerbannog said...

Some followup comments:

Replicating the NASA/NOAA global temperature results *exactly* is a big project. But getting darned close is much easier. You really can get 90+% of the answer with way less than 1% of the effort.

This comes as absolutely no surprise to those who have actually worked with the temperature data. The basics of global temperature computations (and it's only the basics that I've really mastered) just aren't very difficult. These basics could definitely be incorporated into lower-division science/programming coursework at your local university (or even community college).

(An aside: I'd definitely like to encourage college STEM instructors to investigate the possibility of incorporating basic temperature data-crunching exercises into their coursework. Students would have the opportunity to debunk widely-published denier claims about data adjustments, UHI, etc. with software they've written from scratch. That would be both educational and motivational. A plus: having those students go home for the summer and shame their parents into turning off Faux News would definitely be a win.)

The people who have been attacking the NASA/NOAA/BEST/etc. global temperature work range from (numerous) talentless and malicious hacks like Anthony Watts to (much rarer) talented and malicious people like Tony Heller.

Watts doesn't know what the f*&! he's doing, and he doesn't even come close to knowing enough to know that he doesn't know what the f*&! he's doing (convoluted Dunning-Kruger-speak there).

Heller, on the other hand, *must* know exactly what he's doing. He's got more than enough talent to get this stuff right, but he deliberately chooses not to.

The Wattses, Hellers, and their most loyal fans are incapable of feeling shame, but less ideologically rigid people are. A worthy long-range goal would be to limit (and hopefully shrink) the population of people who aren't too embarrassed to be associated with Watts, Heller & their fellow travelers at Heartland, etc.

I've been trying to do that by showing as many people as possible that there's no great mystery about how global-average temperatures are calculated and demonstrating to them that global temperature signal is so strong that it practically "jumps out at you" even with the simplest processing.

Again, this is absolutely *not* news to people who have actually worked with the data. The fundamentals of temperature data processing are so straightforward that reasonable people (when educated) should be too embarrassed to deny it.

caerbannog said...

(Gonna be my own grammar-nazi here)

The fundamentals of temperature data processing are so straightforward that reasonable people (when educated) should be too embarrassed to deny *them*.

Hank Roberts said...

check that first link? not found

The one behind: Here, says Caerbannog pointing to Eli, is proof

caerbannog said...

check that first link? not found

The one behind: Here, says Caerbannog pointing to Eli, is proof

Looks like that tweet got deleted. Here's the tweet of mine that the deleted tweet was referring to:

Johnny Vector said...

Area-weighting? Pish-tosh. Everyone knows area weighting only applies to the presidential popular vote.

Kevin O'Neill said...

caerbannog writes: "Heller, on the other hand, *must* know exactly what he's doing."

This question of why do they get something wrong is the essence of a game I used to play: Ignorant, stupid, insane, or just plain evil? It's a theory that asks why someone gets the wrong answer.

"There is no crime in being ignorant. We are all ignorant on different subjects. I am ignorant on most things Canadian. I don't know how many provinces they have, couldn't name all of them, haven't a clue what the population of each is, etc., etc. You name it -- if it's Canadian -- I probably don't know it. It's nothing to be proud of, but neither is it something of which to be terribly ashamed.
But then, I'm American. They taught us mostly US history in school. If I were a Canadian and didn't know those facts about Canada, I'd have to be pretty stupid. And there are stupid Canadians just as there are stupid Americans and stupid people of every nationality, race, creed, or population grouping of your choice.

Ignorance and stupidity are generally identified by a lack of knowledge. Insanity is a little different. The insane know something is "true" despite overwhelming contrary evidence. It's one thing to be wrong, it's another thing to persist in being wrong when every fact is against you. Often these loonies have conversations inside their heads (delusions) that justify their beliefs. Just completely irrational.

Evil. Knowing right and doing wrong -- intentionally, with eyes wide open. Generally as a result of avarice or a lust for power -- or for the sheer joy of corrupting something good. Evil knows.

As I go through life I play a game. The rules are simple: Observe pseudoskeptics and then identify them correctly as ignorant, stupid, insane or evil."

I'd agree with you on Heller. Watts is a combination of 1 & 2 with a heavy dose of D-K.

Nick Stokes said...

"He's got more than enough talent to get this stuff right"
I don't think so.